CN-122025012-A - Method and system for outlining metabolic tumor volume from time sequence flat scan CT image to guide adaptive radiotherapy based on deep learning

Abstract

The present application relates to a risk assessment method, a risk assessment apparatus, and a computer-readable storage medium. In the risk assessment method, based on a first image (comprising a PET image) acquired by a target object before radiotherapy and a second image acquired during radiotherapy, high-activity area information corresponding to the period of radiotherapy is determined through a high-activity area determination model, and further, a risk assessment result is determined through a risk assessment model based on the high-activity area information, so that a high-activity area of the PET image with higher acquisition cost is mapped through a dynamic image with lower acquisition cost, prognosis prediction is performed according to the high-activity area of the PET image, whether an original plan is to be modified or not is assisted in decision, and the like, thereby realizing a 'closed loop' of accurate medicine, and forming a complete workflow of 'image acquisition- > -artificial intelligent analysis- > risk judgment- > -scheme adjustment'. Therefore, a self-adaptive radiation therapy system integrating perception, prediction and decision can be constructed, and intelligent, accurate and efficient radiation therapy can be realized.

Inventors

• SUN YING
• ZHOU GUANQUN
• HAO MENGYU
• Mo Zijie
• Jia lecheng

Assignees

• 中山大学肿瘤防治中心（中山大学附属肿瘤医院、中山大学肿瘤研究所）
• 上海联影医疗科技股份有限公司

Dates

Publication Date

20260512

Application Date

20260116

Claims (10)

1. 1. A risk assessment method, comprising: Determining high activity area information corresponding to a radiation therapy session by a high activity area determination model based on a first image acquired of a target object prior to the radiation therapy session and a second image acquired during the radiation therapy session, wherein the first image comprises a PET image, and And determining a risk assessment result through a risk assessment model based on the high-activity area information.
2. 2. The risk assessment method according to claim 1, further comprising: and according to the risk assessment result, corresponding radiotherapy clinical decision information is automatically determined.
3. 3. The risk assessment method of claim 1, wherein the high activity region information is determined in real time by the high activity region determination model during the radiation therapy.
4. 4. A risk assessment method according to claim 3, wherein the risk assessment results are determined in real time by the risk assessment model during the radiation therapy.
5. 5. The risk assessment method according to claim 4, further comprising: During the radiation therapy, a treatment plan of the radiation therapy is adjusted in real time according to the risk assessment result.
6. 6. The risk assessment method of claim 1, wherein the high activity region determination model comprises a generative artificial intelligence model.
7. 7. The risk assessment method according to claim 1, wherein the high activity region information comprises delineating data for a high activity region and/or comprises a composite image of a high activity region.
8. 8. The method of risk assessment according to claim 1, wherein, The first image comprises PET/CT image and the second image comprises CT image, or The first image comprises a PET/MR image and the second image comprises an MR image.
9. 9. A risk assessment apparatus, comprising: A region determination module configured to determine high activity region information corresponding to a radiation therapy session by a high activity region determination model based on a first image of a target object acquired prior to the radiation therapy session and a second image acquired during the radiation therapy session, wherein the first image comprises a PET image, and An evaluation module configured to determine a risk assessment result by a risk assessment model based on the high activity region information.
10. 10. A computer readable storage medium having instructions stored thereon, which when executed by a processor implement the risk assessment method of any of claims 1 to 8.

Description

Method and system for outlining metabolic tumor volume from time sequence flat scan CT image to guide adaptive radiotherapy based on deep learning Technical Field The application relates to the technical field of radiotherapy, in particular to a risk assessment method, a risk assessment device and a computer readable storage medium, which can be used for guiding adaptive radiotherapy (also called adaptive radiotherapy) by outlining a metabolic tumor volume from time sequence flat scan CT images based on deep learning. Background In current conventional radiation therapy procedures, there are the following key limitations: 1) The uncertainty of target region delineation is that the delineation of a tumor target region (particularly a PET high metabolic active region) is seriously dependent on the experience of doctors, so that the delineation results among different centers and even different doctors are greatly different, and the curative effect is affected; 2) Treatment response blind areas, namely radiation treatment is usually planned based on static images before treatment, and the biological target area evolution caused by tumor recession, morphological change and functional change among radiation treatment fractions cannot be reflected in real time. This may result in the actual need for an adequate dose of radiation resistant area, or the retracted normal tissue being subjected to unnecessary radiation; 3) The existing prognosis evaluation is mostly based on the characteristic before treatment or follow-up after treatment, and lacks means for early and accurate prediction of curative effect in the radiotherapy process. Intervention in high risk patients cannot be identified at mid-treatment or downgraded treatment is performed for low risk patients to alleviate side effects and economic burden. Therefore, there is a need to develop a method for radiation therapy that addresses the limitations described above. Disclosure of Invention Based on this, it is necessary to provide a risk assessment method, a risk assessment apparatus, and a computer-readable storage medium in order to address the above-described technical problems. In some embodiments, the present disclosure provides a risk assessment method comprising determining, by a high activity area determination model, high activity area information corresponding to a radiation therapy session based on a first image of a target subject acquired prior to the radiation therapy session and a second image acquired during the radiation therapy session, wherein the first image comprises a PET image, and determining, by a risk assessment model, a risk assessment result based on the high activity area information. In some embodiments, the risk assessment method further comprises automatically determining corresponding radiation therapy clinical decision information based on the risk assessment results. In some embodiments, the high activity region information is determined in real time by the high activity region determination model during the radiation therapy. In some embodiments, the risk assessment results are determined in real-time by the risk assessment model during the radiation therapy. In some embodiments, the risk assessment method further comprises adjusting a treatment plan of the radiation therapy in real time during the radiation therapy based on the risk assessment results. In some embodiments, the high activity region determination model comprises a generative artificial intelligence model. In some embodiments, the high activity region information includes delineation data for the high activity region and/or includes a composite image of the high activity region. In some embodiments, the first image comprises a PET/CT image and the second image comprises a CT image, or the first image comprises a PET/MR image and the second image comprises an MR image. In some embodiments, the disclosure provides a risk assessment apparatus comprising a region determination module configured to determine high activity region information corresponding to a radiation therapy session by a high activity region determination model based on a first image of a target subject acquired prior to the radiation therapy session and a second image acquired during the radiation therapy session, wherein the first image comprises a PET image, and an assessment module configured to determine a risk assessment result by a risk assessment model based on the high activity region information. In some embodiments, the present disclosure provides a computer readable storage medium having instructions stored thereon, wherein the instructions, when executed by a processor, implement a risk assessment method as described above. Drawings In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related art, the drawings that are required to be used in the embodiments or the related technical descriptions will be briefly described,